7. Computer controlled machining¶
During this week, I worked on making something big (in the meter scale) on our labs ShopBot CNC machine. I designed, cut, and assemebled something of my own creation, and worked on refining my skills with the ShopBot. In addition to our personal assignment, we were given a group assignment to go through out lab’s safety training and to charecterize our CNC machine
Working on Design¶
The first thing I did this week was figure out what I wanted to design. I thought about it for a while and came up with something that would actually help me: a stool. At home, I have two desks: one for my school work, and another where I put all my fab/engineering things such as my soldering kit and my 3d printer. I usually do all my fab work on this desk, but I have to move a chair from one desk to another whenever I move around. In order to solve this problem, I decided I would design a stool that I can keep at my Fab desk.
I started my first design without looking at any resources or pictures which was a huge mistake as it wasted time. I thought it would be easy and didn’t think too much about it. It was at this point that our lab director, Mr. Tom Dubick told us that we should strive for the extra credit project and create something that is press-fit and uses no glue or fastners.
I went onto Fusion 360 and starting designing away. My initial design consisted of a circle and 4 slits in that circle to insert the four legs into. This is what it looked like
I thought I would test out my design so I used our lab’s laser cutter to cut out a cardboard model of it.
This is how it came out
The design was very impractical and would not stand on its own. It was then I also realized I should be using parameters since I would be scaling the design into the meter scale. Parameters are used to quickly resize selected objects within a design.
I kept trying to fix my design by changeing the shape of the legs
I decided I needed to change my design in order to make it more sturdy. I got rid on the signle outward slits on the seats and replaced them with two parallel slits for each leg. I also added parameters such as material thickness and slot length on the seat.
This is how I designed my second stool and how it turned out.
I thought this design should work. My cardboard cut worked and I had the parameters right for the scaled down version. However, after I scaled the design up, the parameters were not working how I thought they should have. Since I was to be working on the meter scale, I scaled my approximatly 5 inch design by 7x in order to get to close to a meter. With this though, I saw my parameters were not working right and my design was all messed up. No matter how much I changed them, they were not working, which meant I had made an error during my designing. I went back to my sketch and saw that I had not used the parameters like I had intended to, using the wrong ones for the wrong parts of my design.
I decided I had enough of that and looked up pictures of press-fit stools and furniture to get a better understanding of how to tackle the assignment.
I found this image that I thought would work well with what I was thinking of doing
Since our products had to be about a meter long, I decided to make the legs of my design much longer than the one in the picture. I used the image for inspiration on the press-fit parts.
Desiging Final Stool¶
The image inspired what I was going to do for my stool, but I still had to design it from scratch.
I started by creating a sketch of one of the bases for the legs of the stool.
I drew rectangular lines around the basic shape and used the fillet tool to round the corners. After I did that, I went to the top of the design and added the tabs that were going to be pressed into the seat, making it a press-fit stool. Here is where I made another mistkae that costed me some time. I read about dog bone fillets online and thought I should incorporate them into my design. I thought I would add them everywhere there was going to be a press fit piece.
As you can see in this picture below, I spent time on Fusion 360 creating circles at the end of different peices in order to make a dog bone fillet.
This was a huge waste of time as I later learned that Aspire, Vectric CNC’s CAM software, had a tool in it that allowed me to change every corner I needed into a dogbone fillet. However, since I did not know this at the time, I converted all the pressfit peices into dogbone fillets in Fusion 360 and extruded the pieces to get a view of what I had.
I made two copies of the original sketch and layed them over one another in order to get the full view of my stool’s base. To do this, I had to extrude the correct parts on the two different leg sections in order to have them slide into one-another and act as a press-fit construction
The final step in the design process was making the seat for the stool. This was fairly simple since I knew the dimensions of all my slots and I knew the material thickness. I had these as my parameters so creating the seat took no time at all.
This is what my put together model looked like
Preparing Design For CNC¶
After I had my design all ready to go, I had to pull it into Vectric CNC’s Aspire CAM software which we used for our labs ShopBot. I had used Aspire and the ShopBot before, but that was under the surailence of teachers who were willing to help me if I got stuck. Working with Aspire and the ShopBot during Fab Academy was a different story. We were given a basic tutorial about Aspire before the start of the week, but it was still
When I pulled into Aspire, I realized my scale was all messed up It was showing my design as way too big. I realized this was because I was working in mm in Fusion 360, but when I pulled it into aspire, it was converting all the mm into inches so my stool seat was 50 inches rather than 50mm. In order to change this, I had to find the ratio that would scale my project down correctly from 50 inches to 50mm. I figured 50/18 = 100%/x so x had to equal 36. I scaled my project down from 100% to 36% and it was perfect.
Scaled from
To
In addition to scaling my project, I had to add the dog-bone fillets to each corner of my press-fit design. After doing so, I was ready to generate the toolpaths and send it off to the ShopBot to print
Generating Toolpaths In Aspire¶
To generate the toolpaths in Aspire, I started by creating a new file where I was asked to set a workplane size. The workplane of the ShopBot PRSalpha in our lab and where I would be milling was 96” x 48”, and my material height was 0.72”. I then imported my .dxf file from fusion into Aspire. This is when fellow Fab Academy Student Charlie Hovrath asked me if I wanted to put our files together in order to print both of our designs on the same bed while also saving time. Since there are 13 students in working in Fab Academy in our lab, time is of utmost importance to all of us. Charlie and I nested our files together, which worked well becasue we were both eyeing the same sheet of wood. I spent some time moving some of our file around and nesting them snug together in order to make them all fit onto the sheet.
Here is what my layout was before
And this is what it looked like after I merged with Charlie
After we positioned our designs properly and ensured they would fit on the sheet of wood, we started creating toolpaths. We both worked on this seperalty to ensure we knew how to do it by ourselves without the other one watching.
I started with my design. I kept lookign back at my initial Fusion 360 design to determine which parts of each of my bodies would require different toolpath operations. I used both the Profile and the Pocket toolpath for my design, though I could only have used the profile. Profile is used for cutting whole objects out of the wood, while pocket is mostly used for engraving or carving. However, we found a way to use the pocket toolpath to get rid of all the wood in between where we wanted to profile, which saved us time with tabs. I selected all intended profile cuts for each of our files and set the depth of each of these profile cuts to 0.73”, the thickness of the wood. This depth ensured the cuts would go all the way through the material. I then chose the .25 Endmill bit for both my profile and pocket operations. To prevent movement of the wood of each profile during the cuts, I added tabs to to the toolpath in the tabs section of Aspire. These essentially leave a bit of material attached from the wood sheet to the profiles to avoid the profiles from completly losing their place on the bed while the machine is running. The pocket cuts did not require tabs becasue they were completly getting rid of all the material anyways. I also had to make sure the profile cuts had machine vectors running on the outside while the pockets had them running on the inside.
After Charlie did his toolpaths, we saved all of them from Aspire through the Save Toolpath menu. Each of these toolpaths was saved under the Shopbot format, .sbp, and grouped by the type of toolpath it was, pocket or profile. We were officially ready to cut
Running the CNC¶
The first thing we did when cutting on the CNC machine was screw the material sheet down on the machine surface refercing our toolpaths since we did not want our bit running into screws. After putting an adequate number of screws, I was ready to home the axes. Our shopbot comes equipped with proximity switch endstops and a z-probe plate, so homing the axes was fairly easy.
After this machine prep, I was ready to move onto the milling process. Since the operation of teh SHopBot alone can be unsafe, Charlie and I worked on this part together, which was perfect ebcasue both of our designs were on the same files. We started by loading the first tool path, the pocket cut, into the Shopbot Command Console through the Load File menu under the File tab. Under the pop-up menu, 3D Offset was selected for the operations offset. This results in an “air-cut” which allows us to see our cut without tool actually touching the material. Instead, it runs above the bed.
Here is what the aircut looks like
After the aircut, we were ready to cut our pockets. This is how they turned out
After the pockets, we were ready to start the profile cuts. Like the pockets, we ran an air cut to make sure all was good. After that, we started the actual cut.
At this point we thought we were home free. However, the machine had other plans
Broken Bit #1¶
Up to the point in the video above, the profile cut was going good. However, a little after the video ends, a piece of the wood breaks off from the tabs and starts moving around. This causes an error in the machine, and teh bit gets lodged in the sheet of wood.
We had to find a new bit and replace this one that broke. While doing so, we got paranoid that this might happen again, and put in a few more screws in the bed. This was our second mistake.
Broken Bit #2¶
Due to us putting in more screws into the bed, while the machine was runnign the profile cuts the second time, the bit ran into a screw placed in its path.
After this, we made sure to triple check our screws with our toolpaths to make sure none of these errors hapen again. Finnaly, on our third try, we got all the profile cuts finished
Finished Product¶
After we got the cuts finished, I used a wedge to break the tabs that attached my parts to the wood sheet. Since my stool was press-fit, it was easily assembled. I put it together and hammered it a bit to make sure it was together nice and tight.
Outcome¶
Through this week, I really learned how to work with the CNC ShopBot machine after my three tries running it. I refined my skills with CAD design and was delighted to see something so abstract on some software come to frutation. In the near future, I will create something that each leg of my stool will slide into to further its durability.
Group Work¶
In addition to our individual assignments, we were tasked with a group assignment. We were instructed to characterize our CNC ShopBots runout, alignment, fixturing, speeds, feeds, materials, and toolpaths. Our lab has a big ShopBot on which all of us cut our projects on and a smaller CNC. We split ourselves into smaller groups and each group tackled a a different aspect. You can see our work on our group website. I was tasked with measuring alignment for the big ShopBot machine
Files¶
You can find my files for this week here